The Minimum Residual Method of Factor Analysis

Comrey, Andrew L.

doi:10.2466/pr0.11.5.15-18

Cited by 97 publications

(68 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Also referred to as "minres," (minimum residual; Comrey, 1962;Harman, 1967;Harman & Jones, 1966) an OLS solution can be obtained via the widely used computational method known as iterative principal factors (Comrey & Lee, 1992) available in virtually all software packages for factor analysis. It is useful to draw a fundamental distinction between OLS and ML estimation.…”

Section: Ordinary Least Squaresmentioning

confidence: 99%

Recovery of Weak Common Factors by Maximum Likelihood and Ordinary Least Squares Estimation

Briggs¹,

MacCallum²

2003

Multivariate Behavioral Research

214

175

View full text Add to dashboard Cite

This article examines the relative performance of two commonly used methods of parameter estimation in factor analysis, maximum likelihood (ML) and ordinary least squares (OLS). It is shown that ML will sometimes fail to recover a known population factor structure when OLS succeeds. A simulation study was conducted in which two types of error (model and sampling error) were introduced separately and in combination into correlation matrices generated from known population structures with at least one relatively weak major domain factor. Simulated correlation matrices were factor analyzed using both ML and OLS, and recovery of the relatively weak factor(s) was assessed. In situations with a moderate amount of error, ML often failed to recover the weak factor while OLS succeeded. It is suggested that the correspondence between the assumptions inherent in each method regarding error and the actual nature of error in the data may affect the success of recovery of weak common factors. An example using empirical data is also presented.

show abstract

Section: Ordinary Least Squaresmentioning

confidence: 99%

Recovery of Weak Common Factors by Maximum Likelihood and Ordinary Least Squares Estimation

Briggs¹,

MacCallum²

2003

Multivariate Behavioral Research

214

175

View full text Add to dashboard Cite

show abstract

“…In factor analysis, the factor loading under each constraint variable (beta weight) represents a correlation of the variables (constraint areas) to the identified constraint factor and has the same interpretation as any correlation coefficient. However, only variables with loadings of 0.40 and above (10% over lapping variance) were used in naming the factors (Comrey, 1962).…”

Section: Methodsmentioning

confidence: 99%

Challenges of extension workers in reaching rural women farmers in Enugu State Nigeria

Akinnagbe¹,

Ezeuzo²,

Onwubuya³

2018

jae

View full text Add to dashboard Cite

show abstract

“…Items 3,9,10,11,12,14,21,25, and 31 were eliminated from further analysis. The reduced item scale (23 items) was analyzed using Cronbach alpha reliability coefficient (see Anastasi & Urbina, 1997) and Comrey's minimum residual factor analysis (Comrey, 1962;Comrey & Ahumada, 1964) with tandem criteria rotation (Comrey, 1967).…”

Section: Methodsmentioning

confidence: 99%

“…Factor Analysis The minimum residual method of factor extraction (Comrey, 1962) yielded 10 factors. These 10 factors were rotated using Criterion 1 of Comrey's Tandem Criteria of rotation (Comrey, 1967).…”

Section: Reliabilitymentioning

confidence: 99%

Modification and Psychometric Evaluation of the Peterson War Scale

Jones-Wiley¹

2007

View full text Add to dashboard Cite

Summaq~.-A measure on attitudes toward war was administered to 125 student participants at a California university to assess psychometric properties for this scale for possible use in current research. A 5-point scale was substituted for the 2-point one originally. Item analysis indicated 23 of 32 items were viable. Using Cronbach reliability coefficient a and factor analysis, the shortened measure had an internal consistency reliability of 3 5 . Factor analysis yielded a 4-factor structure: (1) War is Bad, (2) War is Necessary, (3) Positive Aspects of War, and (4) No Justification. These results indicate this seeminnlv outdated measure of war attitudes remains useful . , for current research purposes involving measuring attitudes toward war. However, longitudinal research is necessary Recent news article and editorials along with the popularity of blogs and websites3 reflect widespread interest concerning general attitudes toward war. Some reports4 have suggested that the current trend is that of lessened support or favor of war. However, most of these are based on anecdotal information coupled with opinion and informal polls, not empirical scientific research. Sound empirical research needs to be dcne to establish the validity of these apparent trends.Measuring scales are needed to evaluate more accurately the attitudes of people toward war. Some researchers may be tempted to construct a new test for this purpose. This long involved process may not be necessary since

show abstract

The Minimum Residual Method of Factor Analysis

Cited by 97 publications

References 0 publications

Recovery of Weak Common Factors by Maximum Likelihood and Ordinary Least Squares Estimation

Recovery of Weak Common Factors by Maximum Likelihood and Ordinary Least Squares Estimation

Challenges of extension workers in reaching rural women farmers in Enugu State Nigeria

Modification and Psychometric Evaluation of the Peterson War Scale

Contact Info

Product

Resources

About